Camelia, the Perl 6 bug

IRC log for #parrot, 2009-05-01

Parrot | source cross referenced

| Channels | #parrot index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:02 cotto msg bacek I'm working on getting the pmc compiler to emit a proper class_init prototype.  Give me a chance to give up before duplicating the effort. ;)
00:02 purl Message for bacek stored.
00:08 AndyA joined #parrot
00:15 kid51 joined #parrot
00:43 darbelo left #parrot
00:47 bacek_ joined #parrot
00:50 eternaleye joined #parrot
00:52 TonyC joined #parrot
02:21 kid51 planetparrot
02:21 purl somebody said planetparrot was an aggregation of feeds/blogs that are related to parrot (could come from multiple sources) or http://planet.parrotcode.org/
02:25 PacoLinux joined #parrot
02:35 janus joined #parrot
03:06 pmichaud message Tene I think I have most of the .HLL changes in place in master.  Change the definition of RAKUDO_HLL in perl6.pir to switch .HLL settings (throughout the codebase).
03:06 purl Message for tene stored.
03:08 tetragon joined #parrot
04:01 Andy joined #parrot
04:21 davidfetter joined #parrot
06:19 uniejo joined #parrot
06:33 cotto pmichaud, ping
07:00 cxreg interesting discovery about the ruby GC: http://timetobleed.com/6-line-eventmachi​ne-bugfix-2x-faster-gc-1300-requestssec/
07:00 shorten cxreg's url is at http://xrl.us/beq4bp
07:01 cxreg not sure if that has any relevance here
07:07 cotto looks to be of interest, even for non-ruby folks
07:31 cxreg the whole premise of it scouring the stack for things that look like pointers to GCables smells funny, but it's interesting
07:35 iblechbot joined #parrot
07:53 cotto It's a very C solution.
07:56 cotto msg chromatic You a verb in the 3rd paragraph on http://www.modernperlbooks.com/mt/​2009/04/the-why-of-perl-roles.html
07:56 purl Message for chromatic stored.
07:56 shorten cotto's url is at http://xrl.us/beq4eu
07:57 was kicked by cotto: shorten
07:57 cotto I can remember one time when that bot's been useful.
08:06 GeJ when double-clicking a url in irssi doesn't grab anything after (and including) the first '?'... but that would just be me :)
08:09 cotto GeJ, so +1 for not kicking shorten in the future?
08:10 cotto (That sounds like it should be an easy fix, but that's your call.)
08:12 flh joined #parrot
08:12 GeJ Well, I think I could fiddle with my xterm's config. So do as you wish.
08:14 bacek_ joined #parrot
08:38 elmex joined #parrot
08:48 mikehh joined #parrot
09:02 masak joined #parrot
09:37 mikehh after a make smoke i', getting - Could not upload report to Smolder at http://smolder.plusthree.com
09:37 mikehh HTTP CODE: 500 (Internal Server Error)
09:38 mikehh however it has uploaded - http://smolder.plusthree.com/app/pu​blic_projects/report_details/20559
09:39 cotto sounds like some brokenness we had with svn before the move
09:40 bacek good evening
09:40 bacek cotto: hi.
09:41 cotto hi bacek
09:41 mikehh bacek: hi - it;s mid morning for moi
09:41 bacek cotto: VTableInfo.attr stolen from PCT::Node :)
09:41 bacek mikehh: you are living in past :)
09:41 bacek cotto: how class_info emitting going?
09:41 cotto or the FUTUTRE
09:42 mikehh from your POV - probably :-}
09:42 cotto I think I have it implemented, but debugging is confusing
09:42 cotto (as in, I think all the code is there)
09:42 cotto (that needs to be)
09:42 bacek ship it!
09:43 cotto except that it breaks stuff
09:44 mikehh for a while smolder was down - then it came back, but, the upload worked even though it reported errors = the last couple of days
09:45 mikehh I saw somwhere (Freshmeat?) some updates to smp;der
09:45 bacek cotto: break it. We can fix it together.
09:46 bacek cotto: "pmc_pct - highly experimental branch" :)
09:46 mikehh ARRGH - got a new keyboard but it is still OFF - must be my typing
09:47 cotto incoming
09:47 dalek parrot: r38416 | cotto++ | branches/pmc_pct/compilers/pmc (4 files):
09:47 dalek parrot: [pmc_pct] add (not-quite-working) code to emit a class_init function prototype
09:47 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38416/
09:49 bacek rebasing
09:49 purl i heard rebasing was probably the most difficult procedure
09:50 bacek cotto: hey. You committed uncommented "say" in 04-header.t :)
09:51 bacek I did it twice actually :)
09:51 cotto my mistake
09:52 he joined #parrot
09:52 bacek we can clean all this stuff later.
09:53 cotto t/04-header.t breaks.  I'm confused about why it's complaining about an Undef PMC when src/nodes.pir +48 sets class_init.
09:53 cotto s/complaining about/getting/
09:54 cotto (there's some obvious cleanup work to do, since the code only needs to handle one class_init function, but that's for once it's working)
09:56 bacek there is no "class_init" in "class00.pmc".
09:56 bacek May be there is a problem.
09:58 bacek no. it is not.
09:58 cotto yeah, but it should at least be empty
09:59 cotto Also, why does that code (src/nodes.pir around 49) use $I0 without initializing it?
10:02 bacek it shouldn't. Committing now.
10:03 bacek btw. We can always generate class_init in header.
10:04 bacek If there is no user-defined we have to generate "default" in C file.
10:05 bacek incoming
10:05 cotto bacek, of course, so most of that code is unnecessary
10:05 dalek parrot: r38417 | bacek++ | branches/pmc_pct/compilers/pmc/src/nodes.pir:
10:06 dalek parrot: Actually set attributes in PMC::Class.
10:06 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38417/
10:06 cotto (at least for header generation)
10:06 dalek parrot: r38418 | bacek++ | branches/pmc_pct/compilers/​pmc/src/parser/grammar.pg:
10:06 bacek PMC::Class[class_init] can be scalar I think.
10:06 dalek joined #parrot
10:06 cotto You killed him.
10:06 cotto or stunned
10:06 * bacek hides
10:06 cotto You seem to think I implied something which I didn't imply. ;)
10:11 cotto bacek, should I revert that commit, since all headers will need a class_init?
10:11 bacek cotto: let's keep it for now.
10:12 cotto if you say so
10:12 purl damn straight
10:12 cotto karma purl
10:12 purl purl has karma of 8549
10:12 cotto << purl-- >>
10:12 cotto karma purl
10:12 purl purl has karma of 8548
10:12 cotto finally
10:13 bacek cheater!
10:13 bacek :)
10:13 bacek r38419 is "proper" fix for Integer.increment(). We can steal code from Rakudo for proper "join" but it's good for now
10:15 dalek parrot: r38419 | bacek++ | branches/pmc_pct/compilers/pmc/src/builtins.pir:
10:15 dalek parrot: Don't try to join undefined stuff.
10:15 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38419/
10:18 dalek parrot: r38420 | cotto++ | branches/pmc_pct/compilers​/pmc/src/pmc_emitter.pir:
10:18 dalek parrot: [pmc_pct] unconditionally emit a class_init prototype
10:18 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38420/
10:21 cotto bacek, do you mind if I work on making make test work from compilers/pmc ?
10:22 Ron joined #parrot
10:23 bacek cotto: not at all
10:24 bacek I'm thinking about moving generated .pir files into separate directory... I start messing up between sources...
10:25 bacek Hey. What happened to ParrotRoadmap page in trac???
10:27 cotto All the tasks were turned into tickets so we can use Roadmap.
10:45 cotto bacek, committed (although a more proper solution should be found before the branch is merged)
10:45 bacek cotto: agreed
10:46 cotto make it work, then make it work well
10:47 cotto If nobody beats me, I'll look further into it tomorrow.  kid51 will probably be helpful.
11:07 cognominal joined #parrot
11:33 mikehh make -k fulltest at r38415 - make testb, C, f. g. r, S, src-, run-, perl-, benchmark- manifest-, distro-tests PASS
11:34 mikehh codetest, examples-tests FAIL
11:35 mikehh on Kubuntu Jaunty (9.04) AND64
11:40 mikehh nopaste?
11:40 clunker3 http://pasta.test-smoke.org/ or http://paste.husk.org/ or http://nopaste.snit.ch:8001/ or http://rafb.net/paste or http://poundperl.pastebin.com/ or http://paste.scsys.co.uk/
11:40 purl nopaste is at http://nopaste.snit.ch/ (ask TonyC for new channels) or http://rafb.net/paste or http://poundperl.pastebin.com/ or http://paste.scsys.co.uk/ or App::Nopaste or tools/dev/nopaste.pl or at http://www.extpaste.com/ or http://paste.scsys.co.uk (for #catalyst, #dbix-class, #moose  and others)
11:45 nopaste "mikehh" at 90.209.50.184 pasted "fixes for codetest failures (some but not all) at r38415" (71 lines) at http://nopaste.snit.ch/16465
11:48 mikehh the patch fixes trailing space / long line failures in codetest at r38415 - http://nopaste.snit.ch/16465
12:10 mikehh afk for a little while
12:43 Coke anyone else having trouble updating?
12:44 Coke I just opened a ticket with OSU about the non-responsive svn.
12:59 Andy joined #parrot
12:59 Coke Andy: HIO
12:59 Coke Andy: Er, hio.
12:59 Andy hey
12:59 Coke how is sunny Chicago?
13:13 Ron joined #parrot
13:35 gryphon joined #parrot
13:39 mikehh Coke: I don't know if you noticed bot I pasted a patch for some codingstd failures - http://nopaste.snit.ch/16465
13:40 Coke mikehh: I can't do anything about it because svn is dead.
13:40 Coke please post a patch on trac so it doesn't get lost.
13:40 Coke TIA.
13:40 mikehh ok - will do
13:44 mikehh Trac is not responding either!
13:47 Coke whee!
13:48 Coke I presume it will get resolved if/when svn gets resolved.
13:48 Coke osuosl?
14:09 uniejo joined #parrot
14:11 Andy joined #parrot
14:43 Infinoid hmm.  hooray for git
14:43 Infinoid (my mirror is still up if anyone needs a checkout)
15:07 amoc joined #parrot
15:13 dalek parrot: r38421 | cotto++ | branches/pmc_pct/compilers/pmc/t (6 files):
15:13 dalek parrot: [t] make make test work from compilers/pmc, if improperly
15:13 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38421/
15:14 Coke svn is back.
15:18 mikehh so's Trac
15:25 Coke it's google's fault. =-)
15:26 Coke they added a nagios check so this shouldn't go unnoticed again, if it happens.
15:27 mikehh four of the codingstd (trailing spaces) were last modified in r38414 (kid51?) and one (line length) draft/pdd30_install.pod r38384 also
15:27 Theory joined #parrot
15:42 Coke mikehh
15:42 Coke mikehh?
15:42 purl hmmm... mikehh is r36504 - let me svn up etc or working on the darft/pdd's - mikehh's first cut passes t/codingstd/pdd_format.t or getting a TODO passed in t/pmc/exception.t or failing t/pmc/codestring.t - Failed 15/38 subtests on Ubuntu Intrepid Amd64
15:43 mikehh Coke: here
15:43 Coke was looking for your identifying information. just used your nick.
15:44 dalek parrot: r38422 | coke++ | trunk (5 files):
15:44 dalek parrot: Remove trailing whitespace
15:44 dalek parrot: Patch courtesy mikehh++
15:44 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38422/
15:47 mikehh mikehh is Michael Hind or Michael H Hind - H is Holmes as in Sherlock or Mycroft
15:48 mikehh currently residing in Aberdeen in the NE of Scotland
15:49 Infinoid purl, mikehh is also <mailto:mike.hind@gmail.com>
15:49 purl okay, Infinoid.
16:13 gaurav joined #parrot
16:41 HG` joined #parrot
16:42 flh joined #parrot
17:00 Util Current state: `.const .Sub  rsub = 'random_sub'` is invalid syntax,
17:01 Util and            `.const 'Sub' rsub = 'random_sub'` is valid.
17:01 Util Can anyone confirm that this is the intended design? ( `.Sub` bad, `'Sub'` good? )
17:03 jonathan I'm pretty sure .TypeName is dead and burried.
17:03 jonathan (As part of the making type IDs not visible from PIR)
17:04 Util Thanks; I will correct the lingering .TypeName uses.
17:05 jonathan I didn't know that was still parsed.
17:06 Util It is not parsed.
17:06 Util docs/pmc/subs.pod and runtime/parrot/library/OpenGL.pir have it wrong.
17:06 * japhb pops up
17:06 jonathan Ah, OK.
17:07 japhb Util: say what now?  How far back to I need to read?
17:07 jonathan That's good to know.
17:07 Util japhb: just back to "Current state"
17:08 japhb Util: Ah, so I need to fix some code?  Or some docs?
17:08 * japhb goes to update his parrot checkout ....
17:08 japhb Wheee, 800 revs to go ...
17:09 Util japhb: are you the OpenGL person?
17:09 japhb Util: yes.
17:09 japhb Util: I don't get a lot of time to make actual progress, but I try to respond quickly to any problems people find.
17:10 japhb Otherwise, $day_job and @family pretty much own my time.
17:11 japhb Huh ... it looks like someone's been hacking the OpenGL code without me ....
17:12 Util I have OpenGL on my laptop, but I think that no OGL tests are run, but I am still suprised that the .Sub form parses when I do `./parrot runtime/parrot/library/OpenGL.pir`
17:13 japhb I'm guessing it was deprecated but never removed.
17:17 Util But it fails to parse in a simple test program I wrote.
17:17 Coke japhb: did you do a realclean? =-)
17:17 japhb Coke: huhwhat?
17:17 Coke whoops. Util, not japhb
17:18 japhb ah
17:18 Util Coke: Actually, that test is from parrot that Rakudo checks out during --gen-parrot
17:19 rdice joined #parrot
17:20 Coke ah, I see the openGL stuff doing an ack now.
17:20 Coke Util: I get error:imcc:syntax error, unexpected DOT, expecting INTV or FLOATV or STRINGV or
17:21 Coke when trying to use:
17:21 Coke .const .Sub frob = 'glutInit'
17:21 dalek parrot: r38423 | Util++ | trunk/docs/pmc/subs.pod:
17:21 dalek parrot: [docs] Old syntax .const .TYPENAME
17:21 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38423/
17:21 Util Coke, that is what I get on my simple test, but not when running the same ./parrot on the OpenGL file.
17:22 Util Academic, though.
17:22 Coke if something is still allowing the old syntax, it's broken.
17:23 japhb Coke: it looks to me that the old syntax only appears in the POD
17:23 Util I changed the syntax in subs.pod, and will leave OpenGL.pir for japhb.
17:23 Util Losing connection. Be back ~ 5 hours.
17:23 japhb in the OpenGL.pir POD, I mean
17:23 japhb Just an out of date Synopsis.
17:24 Coke japhb: that'd do it. =-)
17:24 dalek parrot: r38424 | japhb++ | trunk/runtime/parrot/library/OpenGL.pir:
17:24 dalek parrot: [OpenGL] Fix old '.const .Sub' syntax in OpenGL.pir POD
17:24 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38424/
17:34 Ron_ joined #parrot
17:55 confound .w 22
18:11 masak joined #parrot
18:13 gravity left #parrot
18:24 Coke anyone here on os x/ppc.
18:24 Coke ?
18:25 Coke or osx 10.5 ?
18:25 purl osx 10.5 is probably supposed to have zfs, iirc
18:25 Coke no, osx 10.5 is leopard.
18:25 purl okay, Coke.
18:31 nopaste joined #parrot
18:44 nopaste joined #parrot
18:45 Infinoid dalek plugins?
18:45 purl i heard dalek plugins was git://squawk.glines.org/botnix-plugins
18:45 Infinoid no, dalek plugins is http://github.com/Infinoid​/dalek-plugins/tree/master
18:45 purl okay, Infinoid.
18:54 rakudohudson joined #parrot
18:59 Casan joined #parrot
19:01 dalek joined #parrot
19:06 Infinoid dalek's lack of rakudo updates may have been as simple as a missing CPAN module on feather.  I'm hoping he'll say something on the next commit.
19:08 jonathan Infinoid: I'll make one soon. :-)
19:25 dalek parrot: r38425 | chromatic++ | trunk/src/pmc/fixedintegerarray.pmc:
19:25 dalek parrot: [PMC] Replaced vtable function calls to find size of fixed integer array with
19:25 dalek parrot: direct C-level attribute lookups.  This provides a modest improvement to
19:25 dalek parrot: PIR function call speed.
19:25 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38425/
19:30 Infinoid jonathan: Thanks, ping me if it doesn't work
19:30 Infinoid (wow, #perl6 is getting big)
19:34 dalek rakudo: e33c004 | jnthn++ | src/builtins/guts.pir:
19:34 dalek rakudo: Implement smart-matching on enum values.
19:34 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/e​33c0049216026a5b3d8855dff0a3ab3bcb983c1
19:34 jonathan Infinoid++
19:42 Coke jonathan: when you refer to a git revision number, that's just a partial, right?
19:43 Infinoid Yeah, the git command accepts the first 6 or 7 digits as a shortcut to the full hash
19:44 Infinoid e.g. "git show e33c004" shows you the commit
19:44 jonathan Coke: yes
19:58 * Coke wonders, if he's the only one who updates darwin entries in PLATFORMS, if he can just rip out darwin stuff. =-)
19:58 Coke (not all of it, just bits)
20:38 dduncan joined #parrot
20:39 Coke Whiteknight--
20:40 dduncan left #parrot
20:41 dalek parrot: r38426 | coke++ | trunk/src/dynext.c:
20:41 dalek parrot: Rerun 'make headerizer' for this file
20:41 dalek parrot: Fixes typo introduced in last commit, passes c_function_docs.t again.
20:41 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38426/
20:45 * Coke fixes 'make headerizer'
20:48 dalek parrot: r38427 | coke++ | trunk (2 files):
20:48 dalek parrot: Fix headerizer directives.
20:48 dalek parrot: Probably broken as a result of recent file moves.
20:48 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38427/
20:59 cghene joined #parrot
21:12 dalek rakudo: e637cef | pmichaud++ | src/builtins/globals.pir:
21:12 dalek rakudo: Get $*IN, $*OUT, $*ERR to default to utf8 encodings.
21:12 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/e​637cef7f50df4157ebae9629737710ba1424c45
21:16 cotton funny what a difference one letter makes
21:27 Whiteknight joined #parrot
21:33 dalek rakudo: 5035385 | jnthn++ | src/classes/Role.pir:
21:33 dalek rakudo: .WHAT on roles should stringify to something with () on the end.
21:33 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/5​035385386acd6a2c8dfe9d6b28ba85e200b4a31
21:33 dalek rakudo: 44c4879 | jnthn++ | src/ (2 files):
21:33 dalek rakudo: Move max to the setting.
21:33 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/4​4c4879faacc50b01fee3682465779a7bcad0d23
21:33 davidfetter joined #parrot
21:35 bacek good morning
21:35 davidfetter hi
21:36 cotto hi
21:37 dalek parrot: r38428 | bacek++ | trunk/compilers/json/postalcodes.pir:
21:37 dalek parrot: Update compilers/json/postalcodes.pir to current Socket API.
21:37 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38428/
22:43 dalek rakudo: ec69e24 | jnthn++ | src/setting/Any-list.pm:
22:43 dalek rakudo: Support arity-1 blocks being supplied to min and max.
22:43 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/e​c69e248f03963aacbde7d54698584c89ec4331c
23:03 Limbic_Region joined #parrot
23:09 dalek parrot: r38429 | bacek++ | branches/pmc_pct (3 files):
23:10 dalek parrot: Rename pmc_emitter.pir to emitter/pmc.pir
23:10 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38429/
23:10 dalek parrot: r38430 | bacek++ | branches/pmc_pct/compilers/pmc (4 files):
23:10 dalek parrot: Refactor C arguments handling
23:10 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38430/
23:13 dalek parrot: r38431 | bacek++ | branches/pmc_pct/compilers/pmc/t (0 files):
23:13 dalek parrot: Reorder tests.
23:13 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38431/
23:20 tetragon joined #parrot
23:27 bobke joined #parrot
23:33 dalek parrot: r38432 | bacek++ | branches/pmc_pct/compilers/pmc (4 files):
23:33 dalek parrot: First cut for emitting C body for VTable methods. Hooray!
23:33 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38432/
23:33 bacek cotto: hooray! r38432 :)
23:43 dalek parrot: r38433 | bacek++ | branches/pmc_pct/compilers/pmc/src (2 files):
23:43 dalek parrot: Rename ret_type to returns in VTableInfo for consistency with PAST::Node
23:43 dalek parrot: review: https://trac.parrot.org/parrot/changeset/38433/
23:46 eternaleye joined #parrot

| Channels | #parrot index | Today | | Search | Google Search | Plain-Text | summary

Parrot | source cross referenced